Testing / comparing : Intel D975XBX2 / Asus P5B DX ***56K WARNING***

I <3 my XBX2 but for reasons that I have no clue about, I can't seem to get my system stable past 2.8GHz.
I have the following:


Intel Core 2 Duo e6300 @ 2.8Hz (for some reason, it won't go faster)
Cooler Master Gemini II HSF w/ 2x 120mm Scythe Minebea NMB fans.
Intel D975XBX2 "Bad Axe 2" Mainboard.
2x 1GB Buffalo FireStix DDR2 PC-8000 ram.
2x Samsung SpinPoint P Series SP2504C 250GB HDs in RAID 0.
2x Samsung SH-S182D/BEBE 18x DVD+/-R/RW DL DVD burners.

Cooling under full load is about 65C and idle is about 40C or so. Anyone have a general idea on what I should be setting cause I am at a lost on how to set it up as I have yet to find an easy to follow guide on the matter.

Quote:

---

Originally Posted by Kurumi

I <3 my XBX2 but for reasons that I have no clue about, I can't seem to get my system stable past 2.8GHz.

---

I have exactly the same situation on my new XBX2. When OC-ed to 2.8 GHz (400 MHz FSB) mbo is full stable, but I just can't get to 430 MHz FSB (3.0GHz). I've tried everything.

Older XBX2 (rev 504) could do 485 MHz FSB without problem, but died last month. New one is rev 507. :(

Nevermind...I'll replace my XBX2 with X38 mbo soon.

Quote:

---

Originally Posted by Kurumi

Anyone have a general idea on what I should be setting cause I am at a lost on how to set it up as I have yet to find an easy to follow guide on the matter.

---

I like the BadAxe2 motherboard. I've not tried to push the FSB as far as you are trying to push it though due to heat (QX6700).

Check these out and see if you can find some answers there:

http://www.peakin.com/xbx2/index.html

http://www.peakin.com/xbx2/calculator.html

Also take a look at this article. Different chipset but you may find something helpful there:

http://www.techspot.com/review/40-co...-overclocking/

anybody have issues with sata ports corrupting/dying..I was getting bads burns..based on burn graph...

On the Intel controller
I had Seagates on Sata 0 and 1....DVDrw on sata 2 and 3(sata2 appears to be the bad port on my mobo...when you use the included sata cables I no longer get the clicking sound when the plug connects

moved the HDD to the marvell controller...burst rate really drops...from 250gb/sec to 173 on my Seagate 320 7200.10

moved the DVDRW to SATA 0 and 1...burns ok again

wonder if I can move my boot drive back to SATA 0 on the intel..leaving a SATA dvdrw on port 1....(place my other sata dvdrw on port3)

I haven't had any issues with my BadAxe2 at all..... No problems with the Intel SATA ports.... No problems with RAM getting fried.... No problems at all that I didn't cause by letting the temps get too high from overclocking further than my Zalman 9700 can handle. I put this computer together in November and have been running is 24/7 crunching WCG since then. At first I had an E6700 in it. Now I have a QX6700 @ 3.3GHz. In addition to crunching WCG 24/7 I use this computer for just about everything else that I do - including making this post.
I'll post all the specs below. Maybe that will help someone out. I don't remember what the voltages are set at. I'll have to come back and edit in that info.

CM Stacker 810 case with two 120mm fans in front instead of one as they come stock. I also added a fan in the side panel - 60mm I think - blowing into the case. Other than the two additional fans the case is stock.

QX 6700
3GB Ballistix Tracer PC2 8000 (2x1GB + 2x512MB)
Zalman 9700 (Ultra 120 Extreme on the way. The 9700 really isn't up to a Quad.)
2 X 150GB Raptors in RAID 0 on the Intel SATA ports + a 320GB Seagate 7200.10
Lite-On DVD
8800 GTX
X-Fi Music
PC P&C Silencer 750

Settings: 330 FSB, RAM 266 / 800 (2/3 ratio)

Volts: CPU 1.3875, FSB 1.375, MCH/ICH 1.550

Hope this helps...... :)

Quote:

---

Originally Posted by David_L6

I haven't had any issues with my BadAxe2 at all..... No problems with the Intel SATA ports.... No problems with RAM getting fried.... No problems at all that I didn't cause by letting the temps get too high from overclocking further than my Zalman 9700 can handle. I put this computer together in November and have been running is 24/7 crunching WCG since then. At first I had an E6700 in it. Now I have a QX6700 @ 3.3GHz. In addition to crunching WCG 24/7 I use this computer for just about everything else that I do - including making this post.
I'll post all the specs below. Maybe that will help someone out. I don't remember what the voltages are set at. I'll have to come back and edit in that info.

CM Stacker 810 case with two 120mm fans in front instead of one as they come stock. I also added a fan in the side panel - 60mm I think - blowing into the case. Other than the two additional fans the case is stock.

QX 6700
3GB Ballistix Tracer PC2 8000 (2x1GB + 2x512MB)
Zalman 9700 (Ultra 120 Extreme on the way. The 9700 really isn't up to a Quad.)
2 X 150GB Raptors in RAID 0 on the Intel SATA ports + a 320GB Seagate 7200.10
Lite-On DVD
8800 GTX
X-Fi Music
PC P&C Silencer 750

Settings: 330 FSB, RAM 266 / 800 (2/3 ratio)

Volts: CPU 1.3875, FSB 1.375, MCH/ICH 1.550

Hope this helps...... :)

---

well I just moved to a new case my TJ07...installed a new dvdrw and all of sudden burns curves are awful...16x disc cant be burned at 16x.....wtf

New dvdrw...swapped cables.....still poop

well move both HDD to marvel off Intel...move SATA to 0 and 1 off SATA 2 & 3....no issues

my sata port is :bananal:

I can't seem to get my Q6600 stable above 3ghz (9x334). It seems my system is more stable with the default fsb and mch voltages (rather than higher ones). Even a touch below default (ie 1.3/1.5) seems fine. There's another post way back in this thread which also indicates success with lowering these voltages. I'll admit, though that my tests have not been the most rigorous (not keeping written notes per gtj's recommendation :-) ). I've booted as high as 360, but stability peak fsb can't be much higher than 334. My stress test consists of running Prime95 25.4 (with 4 threads), ATI Tool, WoW, Coretemp, and IDU. This will cause one of the cores to throw an error pretty quickly if I'm clocked to high. I was hoping to hit around 360, but I guess I won't be too bummed if 334 's the limit.

Quote:

---

Originally Posted by Velatra

I can't seem to get my Q6600 stable above 3ghz (9x334). It seems my system is more stable with the default fsb and mch voltages (rather than higher ones). Even a touch below default (ie 1.3/1.5) seems fine. There's another post way back in this thread which also indicates success with lowering these voltages. I'll admit, though that my tests have not been the most rigorous (not keeping written notes per gtj's recommendation :-) ). I've booted as high as 360, but stability peak fsb can't be much higher than 334. My stress test consists of running Prime95 25.4 (with 4 threads), ATI Tool, WoW, Coretemp, and IDU. This will cause one of the cores to throw an error pretty quickly if I'm clocked to high. I was hoping to hit around 360, but I guess I won't be too bummed if 334 's the limit.

---

are you sure temps are not the issue with loading the quad on air??

also do me a favor try just running Prime 95 with ATI Tool, WOW,AV, spware and IDU(IIRC IDU maybe causing the issues)...drivers maybe conflicting

Hey guys does speedfan report a mobo temp or internal case temp for the badaxe2 mobo

Looks like there's a new v2785 bios update for the 975BX2... but no mention of 6x50 support yet.

Quote:

---

Originally Posted by nealh

...try just running Prime 95 with ATI Tool, WOW,AV, spware and IDU(IIRC IDU maybe causing the issues)...drivers maybe conflicting

---

Well I "primed" successfully with fsb@348 (up from 334) for a few minutes 'til I fired up WoW. This caused Prime95 to throw an error. Even though I was priming successfully prior to launching WoW I still see this as a sign of reduced stability. I like this stress testing method I've hit upon and I think I'll stick with it for the time being. I'll see how I do with the new BIOS. Perhaps I'll also disable VT. Do the rest of y'all disable it? Also do y'all think it's possible that higher voltage settings for FSB and MCH may cause instability (as I seem to find)?

What is the best BIOS for the BadAxe2 right now? I'm trying to run 400x9 stable and it seems that it's having a hard time clocking RAM above 500MHz. I tried 533 with up to 2.28 volts and it would weave in and out of booting; it would work sometimes, sometimes it wouldn't. I know this RAM can go further than 500MHz with 2.2 volts even; I have it running at 500MHz 5-5-5-15 at 2.0v memtest/BioShock stable. :P

Reading back, I read that 2747 seems like a good choice for experimentation. I'm looking for good 24/7 clocks. My system right now:

E6600 333x9 1.35v
2:3 500MHz 5-5-5-15 2.0v
BIOS 2333 (basically everything in my profile)

Also another question. Anyone have problems running 4x1GB setups with quad core? Looking for very modest overclock, along the lines of 333x9 for Q6600 for a friend. Very workstation-y computer for him, and I want it to last so he gets all his friends to buy from me. :P

Quote:

---

Originally Posted by TSFroggy

What is the best BIOS for the BadAxe2 right now? I'm trying to run 400x9 stable and it seems that it's having a hard time clocking RAM above 500MHz.

Anyone have problems running 4x1GB setups with quad core? Looking for very modest overclock, along the lines of 333x9 for Q6600 for a friend. Very workstation-y computer for him, and I want it to last so he gets all his friends to buy from me. :P

---

Bios 2747 is rock solid and is the current reference from 2333. If you have your Watchdog and C1E setting the way you want them then go for the update. 2785 is out now and I am going to try it tonight from 2777.

I will admit that 4 sticks can be a little tricky at times when Overclocking. A couple of times for me the motherboard wouldn't post after some more aggressive OC's, "in my case memory changes." After a lot of experimenting I figured out that the MB has problems letting go of reference frequency changes. When this happened I removed my sticks until it caught back up to the default ram settings. The fix for me was to remove 2 sticks, post and set to default. If that didn't work I brought it down to 1 stick and then posted, from there I just added sticks as I kept reposting sequentially, 1, 2, then 4.

http://downloadcenter.intel.com/Detail_Desc.aspx?agr=N&Inst=Yes&ProductID=2578&Dwn ldID=14267&strOSs=38&OSFullName=OS%20Independent&l ang=eng

New Fixes/Features:
 Sorted the priority of microcodes to be fit into .ROM for
production line verification purpose.

Has anyone found out just exactly what the issue is with the 6850's???? Seems that they work fine in other 975x boards.....:rolleyes:

I can tell you my experience with 6850s on 2 different 975BX2 boards (one a 504 and the other a 507). When I tried bringing them up with 4 1GB sticks of SuperTalent DDR2 800 I got nothing. I was only able to get them to post with 2 sticks of Crucial DDR2 667's. 4 sticks were no go... fans would spin but no post or beeps. I could also get them to post with 2 sticks of the SuperTalent 800's if I forced the bios down to 667 with the Crutials ahead of time. Unfortunately this all took place before I saw this thread. Intel had the 6850's on the BX2 list when I ordered them so I just thought 6850s might be a bad batch and returned them for exchange. Hoping a bios update might fix this but I'm thinking of trying a couple of 2GB 667 sticks as a workaround soon as the replacements show up.

Currently doing Orthos on 3.6GHz with the following settings (30min no error atm):
Reference Frequency: 333MHz
Memory Frequency: 667MHz
Timings: 4-4-4-12-2T and 1:1

vCore: 1.575V -- What application do you guys use to measure vCore under load?
vDimm: 2.16V
vMCH: 1.650V
vFSB: 1.375V

I almost received instant error in Orthos with 1.550V -- How much voltage does other guys with the F-batch require?

I'm still wondering about the difference of using 333 ref 667 mem freq and 266 ref 533 mem freq, am I losing performance by using these options? I would be extremely happy if someone could explain this to me.

Any comment would be appreciated though, I'm quite new into overclocking.

Quote:

---

Originally Posted by KMEI

Currently doing Orthos on 3.6GHz with the following settings (30min no error atm):
Reference Frequency: 333MHz
Memory Frequency: 667MHz
Timings: 4-4-4-12-2T and 1:1

vCore: 1.575V -- What application do you guys use to measure vCore under load?
vDimm: 2.16V
vMCH: 1.650V
vFSB: 1.375V

I almost received instant error in Orthos with 1.550V -- How much voltage does other guys with the F-batch require?

I'm still wondering about the difference of using 333 ref 667 mem freq and 266 ref 533 mem freq, am I losing performance by using these options? I would be extremely happy if someone could explain this to me.

Any comment would be appreciated though, I'm quite new into overclocking.

---

The 333 ref is going to be slower than 266 ref...

I found the 333 ref just not very stable....

I would go to 266:533....also you need very good cooling going over 1.575v on the vcore

I set set vfsb to 1.4v and mch to 1.6v for testing....

I set to 9x 400 fsb with 266 ref and 533.....try to keep vcore at 1.5v or less you are under water...air cooling you are going to get alot of heat

also make in maintenance mode ..watchdog is off....
make sure powerslope is on..this lowers vdroop

I use a digital multimeter to measure vcore at idle and load....

Problem,
I wanted to CPU-Z Validate my OC which required me to upgrade from 1.40 ( I think ) to 1.41. Deleted old CPU-Z map and replaced with new.

Clicked the .exe ....... Computer completely froze, including mouse pointer.
tried to reboot, wouldn't boot anymore. I had to boot into Maintenace menu to discover that mem settings had been reset to SPD (4-4-4-12 1.82V DDR2-800)

This means that CPU-Z forced my computer to run 400 FSB 2:3 and unfortunately my memorys can't handle DDR2-1200 on 1.82V with those timings.

Anyone else having these problems? of course my theory could be wrong but... still the only thing that was touched in the maintenace menu.

vCore: 1.600V BIOS, 1.57V idle, 1.55 load
vFSB: 1.375V
vMCH: 1.600V
vDimm: 2.12V

dunno if I can go any higher, mem won't OC at all...
it doesn't even want to do 430MHz at 5-5-5-15 2.2V

http://img175.imageshack.us/img175/3351/namnlskk7.png

ANOTHER NEW BIOS'

2792

Quite a list of fixes this time.


I have a question.
If i want to OC my Q6600 to Q6850 settings (333FSB 333RAM), will i have to do maintenance mode before any changes to bios or is it only when VCore is changed?
Thats all i wanna try. FSB:RAM is 1:1 atm (266FSB 266RAM)

In case you guys are having problems running CPU-Z 1.41 (to get validations or so) - editing the cpuz.ini to "Sensor=OFF" fixes this problem.

gunnz, it is only if you need to increase voltages yes :)

So i shouldn't buy E6850 because i would not work? I e-mailed intel but, no reply.
Will probably buy Q6600...

Quote:

---

Originally Posted by LaMpiR

So i shouldn't buy E6850 because i would not work? I e-mailed intel but, no reply.
Will probably buy Q6600...

---

No reply to a email?? Now THAT is customer service.....:rolleyes:

http://ve3d.ign.com/articles/news/30...Quad-Core-CPUs

These guys are saying their test system using yorkfield/penryn was running on a bx2 board. How this is so I don't know? Bx2 is not supposed to support these cpus. I still don't even think we have determined even in this thread if the bx2 supports the E6x50 cpus yet alone the new 45nm chips. A lot of speculation but I haven't seen any definitive answers in this thread at all.